A Diagnostic Benchmark for Embodied Spatial Reasoning of Vision-Language Models

Yanpeng Zhao, Wentao Ding, Hongtao Li, Baoxiong Jia, Zilong Zheng

Keywords: spatial cognition, simulated physical world, embodied spatial reasoning, real-to-sim

TL;DR: A diagnostic benchmark for assessing embodied spatial reasoning of vision-language models within a simulated physical environment.

Abstract: A recent trend in vision-language models (VLMs) has been to enhance their spatial cognition for embodied domains. Despite progress, existing evaluations have been limited both in paradigm and in coverage, hindering rapid, iterative model development. To address these limitations, we propose ESPIRE, a diagnostic benchmark for embodied spatial reasoning. ESPIRE offers a simulated world that physically grounds VLMs and evaluates them on spatial-reasoning-centric robotic tasks, thus narrowing the gap between evaluation and real-world deployment. To adapt VLMs to robotic tasks, we decompose each task into localization and execution, and frame both as generative problems, in stark contrast to predominant discriminative evaluations (e.g., via visual-question answering) that rely on distractors and discard execution. This decomposition further enables a fine-grained analysis beyond passive spatial reasoning toward reasoning to act. We systematically design ESPIRE both at the instruction level and at the environment level, ensuring broad coverage of spatial reasoning scenarios. We use ESPIRE to diagnose a range of frontier VLMs and provide in-depth analysis of their spatial reasoning behaviors.

Previously Accepted: Yes

Previous Venue: Accepted to The First Workshop on Efficient Spatial Reasoning at ICLR 2026 (Withdrawn due to disclosure constraints).

Email Sharing: We authorize the sharing of all author emails with Program Chairs.

Data Release: We authorize the release of our submission and author names to the public in the event of acceptance.

Submission Number: 25